This study empirically examines the relation between reportable insider trading and the information captured by annual unexpected earnings for a large sample of firms, spanning a ten-year period (1978--87). Each observation is assigned to one of four groups based on the direction of net insider trading (Buy, Sell) and the sign (+, -) of unexpected earnings. For each of these groups, 15-month cumulative abnormal returns are regressed on annual unexpected earnings. The slope coefficient is the largest for the group where insiders are net purchasers and the sign of unexpected earnings is positive. This is consistent with an inference that insider buying interactively confirms the favorable information captured by positive unexpected earnings and this interaction reduces the noise in unexpected earnings. The result with regard to the unfavorable information captured by the group with insider selling and negative unexpected earnings is similar but less pronounced. The analysis also suggests that insider trading conveys information not fully captured by the year's earnings.